US3990054A - Microprogram organization techniques - Google Patents
Microprogram organization techniques Download PDFInfo
- Publication number
- US3990054A US3990054A US05/521,022 US52102274A US3990054A US 3990054 A US3990054 A US 3990054A US 52102274 A US52102274 A US 52102274A US 3990054 A US3990054 A US 3990054A
- Authority
- US
- United States
- Prior art keywords
- address
- microinstruction
- output
- macroinstruction
- operation code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/22—Microcontrol or microprogram arrangements
- G06F9/26—Address formation of the next micro-instruction ; Microprogram storage or retrieval arrangements
- G06F9/262—Arrangements for next microinstruction selection
- G06F9/264—Microinstruction selection based on results of processing
- G06F9/265—Microinstruction selection based on results of processing by address selection on input of storage
Definitions
- This invention relates to data processing and more particularly relates to apparatus for carrying out a microprogram within a data processor.
- each machine operation is made up of a sequence of microprogram of microinstructions, such as transfers from memory to control or arithmetic registers, or vice-versa, and such as transfers from one register to another.
- the microinstructions are generally stored in a control read-only memory (CROM).
- Control logic enables a microinstruction or a group of microinstructions stored in CROM to execute a program memory instruction or macroinstruction.
- the CROM contents are similar to a group of subroutines which are called by the macroinstructions. Since a different series of microinstructions is stored for each different macroinstruction, the proper entry point to the CROM must be determined by the state of the data processor or the macroinstruction to be executed.
- the operation code of the macroinstruction is used as a direct address of the CROM.
- This technique assumes that a routine in the microprogram is executed from an entry point corresponding to an instruction operation code and that every operation code has a unique entry point. Once an entry point is established, a number of words must be reserved in CROM for instruction execution. This number may be as high as eight, and if the processor uses a seven bit operation code (including the address) the total number of micro words would be 1024 (8 ⁇ 2 7 ). According to this technique, the address of CROM would be represented by the notation (XXXXXX 000) 2 in which the upper seven bits (i.e., the X's) directly correspond to the operation code. Because not all macroinstructions require eight micro words for execution, the CROM would contain blank words and would result in inefficient utilization.
- Another object of the present invention is to provide improved apparatus for quickly and efficiently enabling the microprogram to select the macroinstruction having the highest priority for execution.
- One of the principal purposes of the microprogram is to fetch and execute macroinstructions stored in the program memory of the processor.
- the fetching and executing of some macroinstructions may involve 40-50 microinstructions with various addressing modes, such as direct or indirect.
- Many of the macroinstructions require identical or closely related microinstructions for their execution. If the operation code of the macroinstruction is used as a direct address for CROM, the CROM must store the same microinstructions in connection with each different macroinstruction, thereby resulting in needless duplication and waste of CROM capacity.
- the result can be achieved by dividing the requisite microinstructions into two different levels.
- On the first level the microinstructions relating to the fetching of the operand called for by a macroinstruction are grouped.
- On the second level the microinstructions relating to the arithmetic function applied to the operand are grouped.
- This organization can be achieved by providing a translator which generates two distinct addresses for each operation code stored in the instruction register. The first address points to the first level group of microinstructions and the second address points to the second level group of microinstructions.
- the microinstructions within the group can be sequentially executed by incrementing an address register for the CROM.
- the microinstruction routines requiring branching can be performed by a translator such as a multiple address generator, capable of generating two distinct addresses for each operation code of a macroinstruction being executed.
- a translator such as a multiple address generator, capable of generating two distinct addresses for each operation code of a macroinstruction being executed.
- FIG. 1 is a block diagram schematic drawing of a preferred form of microprogram apparatus made in accordance with the present invention.
- FIG. 2 is a block diagram illustrating the manner in which the microinstructions are organized by the apparatus shown in FIG. 1.
- CROM control read-only memory
- the microprogramming apparatus also includes an instruction register 10 capable of storing a seven bit operation code corresponding to at least a portion of the macroinstruction to be executed.
- the operation code is transmitted over conductors 11-17 to addressing circuitry 18 which selects a particular microinstruction stored in CROM 6 and enables flipflop 7F to read the microinstruction into output buffer 8.
- Addressing circuitry 18 comprises a translating read-only memory (TROM) 19 which takes the form of a diode matrix having eight inputs bits. Seven of the eight input bits are provided over conductors 11-17 and the eighth input bit is provided by a level input conductor 20. TROM 19 translates the operation code stored in register 10 into two different addresses for CROM 6 depending on the state of input 20. The addresses from TROM 19 are transmitted over conductors 21-28.
- TROM translating read-only memory
- Addressing circuitry 18 also comprises a priority encoder 30 which receives input information over input conductors 31-38. If any of conductors 35-38 is switched to a logical one state, the data processor used in connection with the described microprogramming apparatus is said to be in a control panel execution state in which the microprogram stored in CROM 6 is required to perform a specified function. If conductor 34 is switched to a logical one state, the data processor is in an interrupt service state in which the microprogram stored in CROM 6 is expected to perform an interrupt service routine. If conductors 32, 33 are switched to logical one states, the data processor is given direct memory access to CROM 6, that is, the control section of the processor is enabled to perform a direct memory access function between the I/O and memory. If conductor 31 is switched to a logical one state, the processor is in a normal state of operation in which a macroinstruction of the processor program is being fetched from memory and executed in accordance with an operation code.
- Control panel conductors 35-38 have the highest priority
- the interrupt pending conductor 34 has the next highest priority
- direct memory access conductors 32, 33 have the third highest priority
- fetch conductor 31 has the lowest priority.
- Encoder 30 encodes the binary signals appearing on input conductors 31-38 into an eight bit code appearing on output conductors 41-48.
- the bits represented by conductors 44-48 are connected to ground potential and are maintained at a constant zero state.
- the states of conductors 41-43 vary depending on the logical states represented by input conductors 31-38.
- the output conductors from TROM 19 and encoder 30 are connected to a two-to-one multiplexer 50 which transmits the signals appearing on input conductors 21-28 to output conductors 51-58, respectively, if a select input 59 is in a logical zero state and which transmits the signals appearing on input conductors 41-48 to output conductors 51-58, respectively, if select input 59 is in a logical one state.
- Multiplexers of the foregoing type as well-known and can be implemented by standard logic gates.
- the output from multiplexer 50 loads a control read-only memory address register (CROMAR) 60 which is a loadable up-counter having output conductors 61-68 connected to CROM 6.
- a load input 70 loads the signals appearing on conductors 51-58 into register 60, and an increment input 72 increases the numerical value of the signal stored in register 60 by one integer.
- the load/increment functions are exclusive. This is maintained by an inverter 87.
- Inputs 20, 59, 70 and 72 of addressing circuitry 18 are controlled by output stages 81-84 of output buffer register 8.
- Output stages 83 and 84 are connected through zero level select logic circuitry 86 which will be described in more detail later.
- FIG. 2 the circuitry described in FIG. 1 operates as follows.
- the zero level entry into CROM 6 depends on the state of a data processor used in connection with the described microprogramming apparatus. Switching of any one of encoder input lines 31-38 from a zero to a one state transmits a corresponding microinstruction address through multiplexer 50 into CROMAR 60. That is, during the next request for a new address from the priority encoder 30, an address corresponding to an appropriate routine will be passed through multiplexer 50 into CROMAR 60. For example, if conductors 35-38 are switched sequentially from a zero to a one state, an address is entered into CROMAR 60 which will result in the following respective functions:
- conductor 34 is switched from a zero to a one state, an address is generated by encoder 30 and is entered into CROMAR 60 which begins an interrupt service routine stored in CROM 6. Likewise, if either conductor 32 or 33 is switched from a zero to a one state, direct access is provided to CROM 6 from the data processor by performing a memory access function. If conductor 31 is switched from zero to a one state, an address is generated by encoder 30 and is entered into CROMAR 60 which initiates a routine that fetches a macroinstruction to be executed from the main memory of the data processor. Each of the functions represented by conductors 31-38 corresponds to a different routine of microinstructions stored in CROM 6 and each results in a distinct address being loaded into CROMAR 60.
- the instruction fetch routine microinstructions (FIG. 2) are sequentially read out of CROM 6 into output buffer 8.
- the bit located in storage element 81 is a logical zero so that CROMAR 60 is incremented through input 72 at the end of each microinstruction.
- a new microinstruction is sequentially loaded into output buffer 8 until the routine is completed.
- the fetch routine transfers the operation code of the macroinstruction being fetched to register 10 and the fetch routine is completed.
- the bit located in storage element 81 of buffer 8 is a logical one, so that a loading of a new address into CROMAR 60 will be accomplished.
- the bit located in storage element 83 selects the conductors 21-28 as inputs to CROMAR 60 via the 2 to 1 multiplexer 50.
- the bit located in storage element 82 which is also input 20 to TROM 19 is switched to a logical zero state, thus indicating that a first level microinstruction selection is required.
- the value of the microinstruction read into buffer 8 is decoded to determine the level of microinstruction selection.
- the first level selection can follow any one of 6 paths.
- Paths 1 and 2 relate to arithmetic or logical functions requiring two phases: (1) address forming and operand fetching, and (2) execution on the fetched operand.
- the operation code stored in register 10 requires an arithmetic or a logical function, such as addition, subtraction, logical OR or exclusive OR, the operand of the function must be addressed and fetched regardless of the function to be performed.
- the operation code requires different arithmetic functions, such as addition and subtraction, the addressing modes of the operands required for the arithmetic operation may be identical, so that the same operand fetching routine can be used for both arithmetic functions.
- the first level entry address into CROM 6 can be identical for operation codes requiring different arithmetic functions to be performed.
- a second level microinstruction selection is required in order to perform the execution phase.
- different operation codes requiring the same arithmetic function, but using different addressing techniques to obtain the requisite operands can use identical microinstruction routines in order to perform the execution phase.
- the second level entry address into CROM 6 can be identical for operation codes requiring different address forming (level) techniques to obtain the requisite operands. Examples of such operation codes are shown in the following Table 1.
- the last microinstruction in the fetching routine switches select input 20 to a one state, thereby causing TROM 19 to generate second level CROM address 121.
- Address 121 is the first address in the microinstruction routine stored in CROM 6 which performs the add function on the operand fetched by the first level routine beginning at address 023.
- CROMAR 60 is incremented until the last microinstruction of the add routine which completes the execution of the operation code is stored in register 8.
- the last microinstruction enters bits in storage elements 83 and 84 which switch select input 59 into a one state, thereby returning control of the CROM 6 address to encoder 30.
- the concept of using the first and second level CROM addresses generated by TROM 19 can be applied to a wide variety of microinstruction routines.
- the instruction fetch routine is common to all macroinstructions, and its function is to load a new operation code corresponding to the macroinstruction into instruction register 10 and to update the program counter of the data processor (not shown).
- TROM 19 At the end of the fetch routine, TROM 19 generates a first level CROM address corresponding to the operation code stored in register 10. As shown in FIG. 2, the first and second level addressed may follow any one of six major paths which are based on the type of instruction being executed.
- the instructions that take Path One are those having an operand address which is nonindexed.
- the common feature of these instructions is the preparation of the operand based on the addressing mode.
- select input 20 is switched to a one state so that a second level address is generated which reads out the first microinstruction of a routine based on the operations performed.
- select input 59 is switched to a one state so that zero level entry is initiated.
- Instructions which take Path Two are those in which the operand address is indexed. The common quality of these instructions is the address forming routine. Following the execution of the address forming routine, the second level instruction is generated as described in connection with Path One.
- the only instruction which relates to Path Three is a branch unconditional instruction.
- Path Four is followed by conditional branch instructions which must perform a test of an indicator before the program counter can be changed. This function is accomplished by a microinstruction routine represented by the prebranch block, and the results of the test performed by the routine are used to determine the next level of entry (i.e., the next address stored in CROMAR 60). If the test for the indicator is true, a second level address is generated by TROM 19, whereas if the test for the indicator is false, select input 59 is switched to a one state so that a zero level address is initiated.
- Path Five is a catch-all which is followed by all instructions having no common features.
- Path Six is followed by the skip indicator instruction. This instruction is peculiar in the way it handles the program counter and therefore requires a separate path.
- a preskip microinstruction routine represented by the preskip block, tests the indicator. If the indicator test is true, a second level address is generated by TROM 19 which initiates a skip microinstruction routine that increments the program counter. If the indicator test is false, input 59 is switched to a one state so that zero level entry and a zero level address are initiated.
- Zero level select logic 86 controls the operation of select input 59 in the manner indicated in Table 2:
- Table 3 An example set of microinstructions for a digital data computer is shown in Table 3.
- the right hand column of Table 3 entitled PATH indicates the basic instruction paths previously indicated by FIG. 2.
- the bit format of the microinstruction identified in Table 3 is shown in Table 4.
- the definitions of the microinstruction bit shown in Table 4 are provided in Table 5.
- the coding of specific microinstructions and the showing of microcode is shown in Table 6.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Executing Machine-Instructions (AREA)
Abstract
The disclosure describes improved microprogramming apparatus for a data processor. The improved apparatus includes a translating diode matrix which translates the operation code of a macroinstruction being executed into a predetermined address of a microinstruction stored in a read-only memory. For each operation code being executed, the translating diode matrix is capable of reading two different microinstructions held at two different addresses. This technique enables the sharing of microinstructions by macroinstructions requiring common execution techniques. Logic circuitry is also provided to enable conditional branching of the microprogram.
Description
The invention herein described was made in the course of or under a contract or subcontract thereunder with the U.S. Government, Department of the Air Force.
This invention relates to data processing and more particularly relates to apparatus for carrying out a microprogram within a data processor.
In most data processing systems, each machine operation is made up of a sequence of microprogram of microinstructions, such as transfers from memory to control or arithmetic registers, or vice-versa, and such as transfers from one register to another. In these systems, it has been common to use a series or sequence of microinstructions to carry out each machine operation as defined by a macroinstruction. The microinstructions are generally stored in a control read-only memory (CROM). Control logic enables a microinstruction or a group of microinstructions stored in CROM to execute a program memory instruction or macroinstruction. The CROM contents are similar to a group of subroutines which are called by the macroinstructions. Since a different series of microinstructions is stored for each different macroinstruction, the proper entry point to the CROM must be determined by the state of the data processor or the macroinstruction to be executed.
In conventional data processors, the operation code of the macroinstruction is used as a direct address of the CROM. This technique assumes that a routine in the microprogram is executed from an entry point corresponding to an instruction operation code and that every operation code has a unique entry point. Once an entry point is established, a number of words must be reserved in CROM for instruction execution. This number may be as high as eight, and if the processor uses a seven bit operation code (including the address) the total number of micro words would be 1024 (8×27). According to this technique, the address of CROM would be represented by the notation (XXXXXXX 000)2 in which the upper seven bits (i.e., the X's) directly correspond to the operation code. Because not all macroinstructions require eight micro words for execution, the CROM would contain blank words and would result in inefficient utilization.
Accordingly, it is the principal object of the present invention to provide microprogramming apparatus in which blank words in the memory storing the microinstructions are avoided or minimized. The applicant achieves this result by using a translator to convert each operation code to any required address in the CROM.
Since the function of the microprogram is to control the internal architecture of the processor, the microprogram must recognize the current state of the processor and take action corresponding to a particular state. Since come states have a higher priority than others, the microprogram also must be capable of recognizing and selecting the state with the highest priority. Accordingly, another object of the present invention is to provide improved apparatus for quickly and efficiently enabling the microprogram to select the macroinstruction having the highest priority for execution.
One of the principal purposes of the microprogram is to fetch and execute macroinstructions stored in the program memory of the processor. The fetching and executing of some macroinstructions may involve 40-50 microinstructions with various addressing modes, such as direct or indirect. Many of the macroinstructions require identical or closely related microinstructions for their execution. If the operation code of the macroinstruction is used as a direct address for CROM, the CROM must store the same microinstructions in connection with each different macroinstruction, thereby resulting in needless duplication and waste of CROM capacity.
Accordingly, it is another object of the present invention to provide improved apparatus which allows microinstruction sharing during the execution of different macroinstructions. The result can be achieved by dividing the requisite microinstructions into two different levels. On the first level, the microinstructions relating to the fetching of the operand called for by a macroinstruction are grouped. On the second level, the microinstructions relating to the arithmetic function applied to the operand are grouped. This organization can be achieved by providing a translator which generates two distinct addresses for each operation code stored in the instruction register. The first address points to the first level group of microinstructions and the second address points to the second level group of microinstructions. The microinstructions within the group can be sequentially executed by incrementing an address register for the CROM.
According to another feature of the invention, the microinstruction routines requiring branching can be performed by a translator such as a multiple address generator, capable of generating two distinct addresses for each operation code of a macroinstruction being executed.
These and other objects and advantages and features of the present invention will hereafter appear in connection with the accompanying drawings wherein:
FIG. 1 is a block diagram schematic drawing of a preferred form of microprogram apparatus made in accordance with the present invention; and
FIG. 2 is a block diagram illustrating the manner in which the microinstructions are organized by the apparatus shown in FIG. 1.
Referring to FIG. 1, a preferred form of microprogramming apparatus made in accordance with the present invention comprises a control read-only memory (CROM) 6 which stores microinstructions at predetermined addresses and reads out the macroinstructions on demand into an output buffer register 8 in response to a read out signal applied to a conductor 7 by a control flipflop 7F.
The microprogramming apparatus also includes an instruction register 10 capable of storing a seven bit operation code corresponding to at least a portion of the macroinstruction to be executed. The operation code is transmitted over conductors 11-17 to addressing circuitry 18 which selects a particular microinstruction stored in CROM 6 and enables flipflop 7F to read the microinstruction into output buffer 8.
Addressing circuitry 18 comprises a translating read-only memory (TROM) 19 which takes the form of a diode matrix having eight inputs bits. Seven of the eight input bits are provided over conductors 11-17 and the eighth input bit is provided by a level input conductor 20. TROM 19 translates the operation code stored in register 10 into two different addresses for CROM 6 depending on the state of input 20. The addresses from TROM 19 are transmitted over conductors 21-28.
Addressing circuitry 18 also comprises a priority encoder 30 which receives input information over input conductors 31-38. If any of conductors 35-38 is switched to a logical one state, the data processor used in connection with the described microprogramming apparatus is said to be in a control panel execution state in which the microprogram stored in CROM 6 is required to perform a specified function. If conductor 34 is switched to a logical one state, the data processor is in an interrupt service state in which the microprogram stored in CROM 6 is expected to perform an interrupt service routine. If conductors 32, 33 are switched to logical one states, the data processor is given direct memory access to CROM 6, that is, the control section of the processor is enabled to perform a direct memory access function between the I/O and memory. If conductor 31 is switched to a logical one state, the processor is in a normal state of operation in which a macroinstruction of the processor program is being fetched from memory and executed in accordance with an operation code.
The foregoing states have a priority associated with their occurrence and are treated as priority coded interrupts to the microprogramming apparatus. Control panel conductors 35-38 have the highest priority, the interrupt pending conductor 34 has the next highest priority, direct memory access conductors 32, 33 have the third highest priority and fetch conductor 31 has the lowest priority.
Encoder 30 encodes the binary signals appearing on input conductors 31-38 into an eight bit code appearing on output conductors 41-48. The bits represented by conductors 44-48 are connected to ground potential and are maintained at a constant zero state. The states of conductors 41-43 vary depending on the logical states represented by input conductors 31-38.
The output conductors from TROM 19 and encoder 30 are connected to a two-to-one multiplexer 50 which transmits the signals appearing on input conductors 21-28 to output conductors 51-58, respectively, if a select input 59 is in a logical zero state and which transmits the signals appearing on input conductors 41-48 to output conductors 51-58, respectively, if select input 59 is in a logical one state. Multiplexers of the foregoing type as well-known and can be implemented by standard logic gates.
The output from multiplexer 50 loads a control read-only memory address register (CROMAR) 60 which is a loadable up-counter having output conductors 61-68 connected to CROM 6. A load input 70 loads the signals appearing on conductors 51-58 into register 60, and an increment input 72 increases the numerical value of the signal stored in register 60 by one integer. The load/increment functions are exclusive. This is maintained by an inverter 87.
Referring to FIG. 2, the circuitry described in FIG. 1 operates as follows.
The zero level entry into CROM 6 depends on the state of a data processor used in connection with the described microprogramming apparatus. Switching of any one of encoder input lines 31-38 from a zero to a one state transmits a corresponding microinstruction address through multiplexer 50 into CROMAR 60. That is, during the next request for a new address from the priority encoder 30, an address corresponding to an appropriate routine will be passed through multiplexer 50 into CROMAR 60. For example, if conductors 35-38 are switched sequentially from a zero to a one state, an address is entered into CROMAR 60 which will result in the following respective functions:
(1) writing to memory from the control panel (conductor 35);
(2) displaying from memory under instructions from the control panel (conductor 36);
(3) executing an instruction from the control panel (conductor 37); and
(4) loading the accumulator from the panel (conductor 38).
If conductor 34 is switched from a zero to a one state, an address is generated by encoder 30 and is entered into CROMAR 60 which begins an interrupt service routine stored in CROM 6. Likewise, if either conductor 32 or 33 is switched from a zero to a one state, direct access is provided to CROM 6 from the data processor by performing a memory access function. If conductor 31 is switched from zero to a one state, an address is generated by encoder 30 and is entered into CROMAR 60 which initiates a routine that fetches a macroinstruction to be executed from the main memory of the data processor. Each of the functions represented by conductors 31-38 corresponds to a different routine of microinstructions stored in CROM 6 and each results in a distinct address being loaded into CROMAR 60.
Assuming conductor 31 is switched from a zero to a one state, the instruction fetch routine microinstructions (FIG. 2) are sequentially read out of CROM 6 into output buffer 8. In each microinstruction of the fetch routine, the bit located in storage element 81 is a logical zero so that CROMAR 60 is incremented through input 72 at the end of each microinstruction. As a result, a new microinstruction is sequentially loaded into output buffer 8 until the routine is completed. The fetch routine transfers the operation code of the macroinstruction being fetched to register 10 and the fetch routine is completed. In the last microinstruction of the fetch routine, the bit located in storage element 81 of buffer 8 is a logical one, so that a loading of a new address into CROMAR 60 will be accomplished. At the same time, the bit located in storage element 83 selects the conductors 21-28 as inputs to CROMAR 60 via the 2 to 1 multiplexer 50. The bit located in storage element 82 which is also input 20 to TROM 19 is switched to a logical zero state, thus indicating that a first level microinstruction selection is required. In summary, the value of the microinstruction read into buffer 8 is decoded to determine the level of microinstruction selection.
As shown in FIG. 2, the first level selection can follow any one of 6 paths. Paths 1 and 2 relate to arithmetic or logical functions requiring two phases: (1) address forming and operand fetching, and (2) execution on the fetched operand. If the operation code stored in register 10 requires an arithmetic or a logical function, such as addition, subtraction, logical OR or exclusive OR, the operand of the function must be addressed and fetched regardless of the function to be performed. Even if the operation code requires different arithmetic functions, such as addition and subtraction, the addressing modes of the operands required for the arithmetic operation may be identical, so that the same operand fetching routine can be used for both arithmetic functions. In such a case, the first level entry address into CROM 6 can be identical for operation codes requiring different arithmetic functions to be performed.
When the address forming phase is completed, a second level microinstruction selection is required in order to perform the execution phase. In some cases, different operation codes requiring the same arithmetic function, but using different addressing techniques to obtain the requisite operands, can use identical microinstruction routines in order to perform the execution phase. In such cases, the second level entry address into CROM 6 can be identical for operation codes requiring different address forming (level) techniques to obtain the requisite operands. Examples of such operation codes are shown in the following Table 1.
TABLE 1
__________________________________________________________________________
Level Entry Example
Required Operation Code
Operation
1st Level
2nd Level
Function Code CROM Address
CROM Address
__________________________________________________________________________
ADD Direct, page zero
(010).sub.8
(023).sub.8
(121).sub.8
ADD Direct, PC Relative
(011).sub.8
(027).sub.8
(121).sub.8
ADD Indirect, page zero
(012).sub.8
(031).sub.8
(121).sub.8
ADD Indirect, PC Relative
(013).sub.8
(035).sub.8
(121).sub.8
SUB Direct, page zero
(020).sub.8
(023).sub.8
(125).sub.8
SUB Direct, PC Relative
(021).sub.8
(027).sub.8
(125).sub.8
SUB Indirect, page zero
(022).sub.8
(031).sub.8
(125).sub.8
SUB Indirect, PC Relative
(023).sub.8
(035).sub.8
(125).sub.8
__________________________________________________________________________
Referring to Table 1, if operation code 010 is stored in registerr 10, an add function is to be performed on the operand which is directly addressed at page zero of main memory. As previously explained, the last microinstruction of the fetch routine switches input 20 to a zero state. As a result, TROM 19 produces a first level CROM address 023 which is transmitted through multiplexer 50 and stored in CROMAR 60. Address 023 is the address of the first microinstruction in a routine stored in CROM 6 for fetching the operand from page zero of main memory. Additional microinstructions in the fetching routine are read out into output buffer 8 by incrementing CROMAR 60. The last microinstruction in the fetching routine switches select input 20 to a one state, thereby causing TROM 19 to generate second level CROM address 121. Address 121 is the first address in the microinstruction routine stored in CROM 6 which performs the add function on the operand fetched by the first level routine beginning at address 023. In order to perform the add function, CROMAR 60 is incremented until the last microinstruction of the add routine which completes the execution of the operation code is stored in register 8. The last microinstruction enters bits in storage elements 83 and 84 which switch select input 59 into a one state, thereby returning control of the CROM 6 address to encoder 30.
As shown in FIG. 2, the concept of using the first and second level CROM addresses generated by TROM 19 can be applied to a wide variety of microinstruction routines. As previously pointed out, the instruction fetch routine is common to all macroinstructions, and its function is to load a new operation code corresponding to the macroinstruction into instruction register 10 and to update the program counter of the data processor (not shown).
At the end of the fetch routine, TROM 19 generates a first level CROM address corresponding to the operation code stored in register 10. As shown in FIG. 2, the first and second level addressed may follow any one of six major paths which are based on the type of instruction being executed.
Path One
The instructions that take Path One are those having an operand address which is nonindexed. The common feature of these instructions is the preparation of the operand based on the addressing mode. After completion of the address forming, select input 20 is switched to a one state so that a second level address is generated which reads out the first microinstruction of a routine based on the operations performed. After the second level routine is completed, select input 59 is switched to a one state so that zero level entry is initiated.
Path Two
Instructions which take Path Two are those in which the operand address is indexed. The common quality of these instructions is the address forming routine. Following the execution of the address forming routine, the second level instruction is generated as described in connection with Path One.
Path Three
The only instruction which relates to Path Three is a branch unconditional instruction. There are four different routines within the branch routine program change counter block because the different addressing modes, but the final result of all the routines is a change in the program counter of the data processor (not shown).
Path Four
Path Four is followed by conditional branch instructions which must perform a test of an indicator before the program counter can be changed. This function is accomplished by a microinstruction routine represented by the prebranch block, and the results of the test performed by the routine are used to determine the next level of entry (i.e., the next address stored in CROMAR 60). If the test for the indicator is true, a second level address is generated by TROM 19, whereas if the test for the indicator is false, select input 59 is switched to a one state so that a zero level address is initiated.
Path Five
Path Five is a catch-all which is followed by all instructions having no common features.
Path Six
Path Six is followed by the skip indicator instruction. This instruction is peculiar in the way it handles the program counter and therefore requires a separate path. A preskip microinstruction routine, represented by the preskip block, tests the indicator. If the indicator test is true, a second level address is generated by TROM 19 which initiates a skip microinstruction routine that increments the program counter. If the indicator test is false, input 59 is switched to a one state so that zero level entry and a zero level address are initiated.
Generation of the zero level, first level and second level addresses is controlled by inputs 20, 59 and 72. These inputs, in turn, are controlled by storage elements 81-84 in output buffer 8 and by zero level select logic 86. Zero level select logic 86 controls the operation of select input 59 in the manner indicated in Table 2:
TABLE 2
__________________________________________________________________________
Output
Output
Output
Output
State of
State of
State of
State of
Element
Element
Element
Element
Input
Input
Input
Input
81 82 83 84 70 72 20 59 Function Performed
__________________________________________________________________________
0 -- -- -- 0 1 -- -- Increment CROMAR 60 (next
sequential microinstruction
to be executed)
1 -- 1 0 1 0 -- 1 Load CROMAR 60 with zero
level address
1 0 0 0 1 0 0 0 Load CROMAR 60 with first
level address
1 1 0 0 1 0 1 0 Load CROMAR 60 with second
level address
1 0 or 1
0 1 1 0 0 or 1
1 (if test
Load CROMAR 60 with zero level
is false)
address if test is false,
otherwise load CROMAR 60 with
address having level specified
by state of TROM input
__________________________________________________________________________
20
The design of such logic would be obvious to those skilled in the art based on the information given in Table 2. As shown in Table 2, the last four bits of each microinstruction stored in output buffer 8 (which controls storage elements 81-84), provided all the information necessary to achieve the zero, first, and second level addresses indicated in FIG. 2. One advantage of this control technique is that branching does not require address storage in the microinstructions, but depends only on the proper switching of inputs 20, 59, 70 and 72 to transmit the proper level address to CROMAR 60.
An example set of microinstructions for a digital data computer is shown in Table 3. The right hand column of Table 3 entitled PATH indicates the basic instruction paths previously indicated by FIG. 2. The bit format of the microinstruction identified in Table 3 is shown in Table 4. The definitions of the microinstruction bit shown in Table 4 are provided in Table 5. The coding of specific microinstructions and the showing of microcode is shown in Table 6.
TABLE 3
__________________________________________________________________________
EXAMPLE INSTRUCTION SET
INSTRUCTION INSTRUCTION BIT COND PA
MNEMONIC
FUNCTION 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
CODES
TH
__________________________________________________________________________
WFI WAIT FOR INTERRUPT 5
LDA LOAD ACCUMULATOR FROM MEMORY P,N,Z
1
STA STORE ACCUMULATOR IN MEMORY 1
ADD ADD MEMORY TO ACCUMULATOR C,P,N,Z
1
SUB SUBTRACT MEMORY FROM ACCUMULATOR C,P,N,Z
1
LOR LOGICAL OR MEMORY AND ACCUMULATOR P,N,Z
1
XOR EXCLUSIVE OR MEMORY AND ACCUMULATOR P,N,Z
1
AND LOGICAL AND MEMORY AND ACCUMULATOR P,N,Z
1
JMP JUMP UNCONDITIONAL 3
JTS JUMP TO SUBROUTINE 4
JTI JUMP TO INTERRUPT SUBROUTINE 4
BON BRANCH ON NEGATIVE ACCUMULATOR
OP CODE DISPLACEMENT 4
BOP BRANCH ON POSITIVE ACCUMULATOR 4
BOZ BRANCH ON ZERO INDICATOR 4
INC INCREMENT MEMORY 1
LCA LOAD CPU ADDRESS FOR DMA 1
LWC LOAD WORD COUNT FOR DMA 1
LIR LOAD IOP ADDRESS-TRANSFER TO IOP 1
LIW LOAD IOP ADDRESS-TRANSFER TO CPU 1
DIB DECREMENT INDEX, BRANCH IF NOT ZERO 4
LDI LOAD ACCUMULATOR - POST INDEXED P,N,Z,
2
STI STORE ACCUMULATOR - POST INDEXED 2
ADI ADD TO ACCUMULATOR - POST INDEXED C,P,N,Z
2
JMI JUMP UNCONDITIONAL - POST INDEXED 5
NOP NO OPERATION 5
SBM0 SUBSTITUTE BYTE 0 IN MEMORY 5
SBM1 SUBSTITUTE BYTE 1 IN MEMORY 5
SBM2 SUBSTITUTE BYTE 2 IN MEMORY 5
SBM3 SUBSTITUTE BYTE 3 IN MEMORY 5
SBA0 SUBSTITUTE BYTE 0 IN ACCUMULATOR P,N,Z
5
SBA1 SUBSTITUTE BYTE 1 IN ACCUMULATOR P,N,Z
5
SBA2 SUBSTITUTE BYTE 2 IN ACCUMULATOR P,N,Z
5
SBA3 SUBSTITUTE BYTE 3 IN ACCUMULATOR P,N,Z
5
SAR SHIFT ARITHMETIC RIGHT P,N,Z
5
SAL SHIFT ARITHMETIC LEFT P,N,Z
5
SLR SHIFT LOGICAL RIGHT NO. OF SHIFTS
P,N,Z
5
SLL SHIFT LOGICAL LEFT P,N,Z
5
SRR SHIFT ROTATE RIGHT P,N,Z
5
SRL SHIFT ROTATE LEFT P,N,Z
5
SBT SINGLE BIT TEST BIT Z 5
SBS SINGLE BIT SET POSITION
P,N,Z
5
SBR SINGLE BIT RESET P,N,Z
5
XIN EXECUTE INPUT ADDRESS
FUNCTION
P,N,Z
5
XOT EXECUTE OUTPUT 5
CTA CONTROL MEMORY TO ACCUMULATOR CM P,N,Z
5
ATC ACCUMULATOR TO CONTROL MEMORY ADDRESS 5
LII LOAD INDEX IMMEDIATE CONSTANT 5
SIN SKIP IF INDICATOR NOT SET INDICATORS 6
CLA CLEAR ACCUMULATOR P,N,Z
5
OCA ONE'S COMPLEMENT ACCUMULATOR P,N,Z
5
COM TWO'S COMPLEMENT ACCUMULATOR P,N,Z
5
RFS RETURN FROM SUBROUTINE 5
RFI RETURN FROM INTERRUPT SUBROUTINE 5
EIN ENABLE INTERRUPTS 5
DIN DISABLE INTERRUPTS 5
__________________________________________________________________________
TABLE 4 ______________________________________ MICROINSTRUCTION BIT FORMAT ______________________________________SHIFT ACC MM 2 × 1 CTRL ALU 1/2 WORD ACC ALU CODE DMAI ALU CODE CST ALU CODE IN ALU CODE INTR ALU MODE CM CARRY IN MAR C.C.STORE A C.C. RESTORE B C.C. SAMPLE CM DIS. INTR. ACC EN. INTR. ACCB WAIT INTR. MM DMA READ WRC DMA WRITE CPU AR WAIT DMA DMAO I/O STROBE OUT CLEAR A FAR CLEAR B IR CM CODE LEVEL CTRL CM CODE LEVEL CTRL CM CODE LEVEL CTRL CM CODE LEVEL CTRL ______________________________________
TABLE 5
__________________________________________________________________________
MICROINSTRUCTION BIT DEFINITION
Bit Function
__________________________________________________________________________
0 MM Enable memory data to the bus
1 ALU Enable arithmetic and logic unit output
to the bus
2 ACC Enable accumulator data to the bus
3 DMAI Enable DMA input data to the bus
4 CST Enable constant generator data to the bus
5 IN Enable I/O input data to the bus
6 INTR Enable interrupt address to the bus
7 CM Enable control memory data to the bus
8 MAR Load memory address register with bus data
9 A Load A register in arithmetic and logic
unit with bus data
10 B Load B register in arithmetic and logic
unit with bus data
11 CM Load control memory with bus data
12 ACC Load accumulator with bus data
13 ACCB Load accumulator with bus data, use byte
load control
14 MM Load memory with bus data
15 WCR Load word count register with bus data
16 CPUAR Load CPU address register with bus data
17 DMAO Load DMA output register with bus data
18 OUT Enable bus to I/O output bus
19 FAR Load function and address register with
bus data
20 IR Load instruction register with bus data
21-24
CM code 0 use instruction bits 7, 8, 9 to address
control memory
000 program counter
001 Index register
010 Pointer register
011 Base register
100 Temp A
101 Temp B
110 Temp C
111 Temp D
25 SPARE
26 SHIFT ACC
Perform shift cycle in the accumulator
27 2×1 CTRL
Concatenate A and B registers in ALU
28 1/2 word
Configure the ALU for a 16 bit operation
29-33
ALU code
Function to be performed by the ALU
ALU mode
ex., add, and, or ......
34 Carry IN
Carry in condition to the ALU
35 CC STORE
Store condition codes in the shadow register
36 CC Restore
Restore condition codes from the shadow
register
37 CC SAMPLE
Sample condition of data on the bus and
update condition codes
38 DIS INTR
Disable interrupts
39 EN INTR Enable interrupts
40 WAIT INTR
Wait for interrupt, halt execution of micro-
instructions until interrupt occurs
41 DMA READ
Initiate DMA memory read
42 DMA WRITE
Initiate DMA memory write
43 WAIT DMA
Wait for DMA Memory transfer
44 I/O STROBE
Generate I/O strobe
45 Clear A Clear A register in ALU
46 Clear B Clear B register in ALU
47 SPARE
48-51
LEVEL CTRL
Level control
__________________________________________________________________________
TABLE 6
__________________________________________________________________________
MICROCODE (1)
C.M. ALU Level
Source
Destination
Address Code Control
__________________________________________________________________________
Zero Level Entry
FETCH CM MAR,A PC Increment
MM IR,B Increment
ALU CM PC 1 F=A A 1 First
INTERRUPT FETCH
INTR
MAR,A Increment
MM IR,B First
Address Forming
Non-Indexed
DIRECT-BASE
ALU MAR 1 A
1 F=B L Second
DIRECT-P.C. REL
ALU MAR A
1 F=B L Second
IND-BASE ALU MAR 1 A
1 F=B L Increment
MM B Increment
ALU MAR B
1 F=B L Second
IND-P.C. REL
ALU MAR A
1 F=B L Increment
MM B Increment
ALU MAR B
1 F=B L Second
Instruction
Execution-Group 1
LD MM ACC 1 Zero
TA ACC MM Zero
ADD MM B Increment
ACC A Increment
ALU ACC B
0 F=A plus B
A 0 1 Zero
SUB MM B Increment
ACC A Increment
ALU ACC B
0 F=A minus B-1
A 1 1 Zero
LOR MM B Increment
ACC A Increment
ALU ACC B
0 F=A+B L 1 Zero
XOR MM B Increment
ACC A Increment
ALU ACC B
0 F=A + B L 1 Zero
MICROCODE (2)
AND MM B Increment
ACC A Increment
ALU ACC B
0 F=A.B L 1 Zero
INC MM B 1 Increment
ALU MM B
0 F=A plus B
A
1 Zero
LCA MM CPUAR Zero
LWC MM WCR Zero
LIR MM CM TEMP 1 Increment
REENTER CM MAR,A TEMP 1 Increment
ALU CM TEMP B
1 F=A plus B
A
1 1 Increment
DMAI
MM 1 Zero
LIW MM CM TEMP Increment
CM MAR,A TEMP 1 Increment
ALU CM TEMP B
1 F=A plus B
A
1 Increment
MM DMAO Increment
REENTER CM MAR,A TEMP 1 1 Increment
ALU CM TEMP B
1 F=A plus B
A
1 Increment
MM DMAO 1 Zero
Address Forming
Indexed
DIRECT-BASE
ALU CM TEMP 1 A
1 F=B L Increment
CM A TEMP Increment
CM B INDEX Increment
ALU MAR B
1 F=A plus B
A
0 Second
DIRECT-P.C. REL
ALU CM TEMP A
1 F=B L Increment
CM A TEMP Increment
CM B INDEX Increment
ALU MAR B
1 F=A plus B
A
0 Second
IND-BASE ALU MAR 1 A
1 F=B L Increment
MM A Increment
CM B INDEX Increment
ALU MAR B
1 F=A plus B
A
0 Second
NOTE:
The ALU field is used to specify the test. The test logic is activated
from the level control.
MICROCODE (3)
IND-P.C. REL
ALU MAR A
1 F=B L Increment
MM A Increment
CM B INDEX Increment
ALU MAR B
1 F=A plus B
A
0 Second
Branch
DIRECT-BASE
ALU CM PC 1 A
1 F=B L Zero
DIRECT-P.C. REL
ALU CM PC A
1 F=B L Zero
IND-BASE ALU MAR 1 A
1 F=B L Increment
MM CM PC Zero
IND-P.C. REL
ALU MAR A
1 F=B L Increment
MM CM PC Zero
Pre-Branch
JTS ALU CM TEMP 1 F=A L Increment
CM A PTR Increment
ALU MAR,CM PTR 1 F=A A
1 Increment
CM MM PC Increment
CM A TEMP Second
JTI ALU CM TEMP 1 F=A L 1 1 Increment
CM A PTR Increment
ALU MAR,CM PTR 1 F=A A
1 Increment
CM MM PC Increment
CM A TEMP Second
DIB ALU CM TEMP 1 F=A L Increment
CM A INDEX Increment
ALU CM INDEX 1 F=A minus 1
A
0 1 Increment
CM A TEMP TEST O Condition
Second
BON TEST N Condition
Second
BOP TEST P Condition
Second
BOZ TEST O Condition
Second
MICROCODE (4)
Instruction
Execution-Group 2
JMI-DIR BASE
ALU CM TEMP 1 A
1 F=B L Increment
CM A TEMP Increment
CM B INDEX Increment
ALU CM PC B
1 F=A plus B
A
0 Zero
JMI-DIR P.C.REL
ALU CM TEMP A
1 F=B L Increment
CM A TEMP Increment
CM B INDEX Increment
ALU CM PC B
1 F=A plus B
A
0 Zero
JMI-IND BASE
ALU MAR 1 A
1 F=B L Increment
MM A Increment
CM B INDEX Increment
ALU CM PC B
1 F=A plus B
A
0 Zero
JMI-Ind P.C.REL
ALU MAR A
1 F=B L Increment
MM A Increment
CM B INDEX Increment
ALU CM PC B
1 F=A plus B
A
0 Zero
NOP Zero
SBA ALU MAR 1 A
1 F=B L
0 Increment
MM ACC B Increment
- ACC 1 Zer
o
SBM ALU MAR 1 A
1 F=B L Increment
ACC A Increment
MM ACC Increment
ALU ACC B Increment
ACC MM Increment
ALU ACC F=A L Zero
SAR, SAL
SRL, SLL
SRR, SLR ACC 1 1 Zero
MICROCODE (5)
SBT ACC A Increment
CST B Increment
ALU B F=A.B L 1 Zero
SBS ACC A Increment
CST B Increment
ALU ACC B F=A+B L 1 Zero
SBR ACC A Increment
CST B Increment
ALU ACC B F=A.B L 1 Zero
XIN ALU FAR B F=B L Increment
Increment
Increment
IN ACC 1 1 Increment
1 1 Increment
ALU FAR F=A L Zero
XOT ALU FAR B F=B L Increment
ACC OUT Increment
Increment
Increment
1 Increment
1 1 Increment
ALU FAR F=A L Zero
NOTE:
The above instructions contain blank words because the I/O timing has not
been finalized.
CTA CM ACC USE 7,8 1 Zero
ATC ACC CM USE 7,8 Zero
LII ALU CM INDEX 1 A F=B L Zero
CLA ALU ACC 1 F=A L Zero
OCA ACC A Increment
ALU ACC F=A L 1 Zero
COM ACC B 1 Increment
ALU ACC B OF=A minus B-1
A 1 1 Zero
MICROCODE (6)
WFI 1 Zero
RFS CM MAR,A PTR Increment
ALU CM PTR 1 F=A minus 1
A
0 Increment
MM CM PC Zero
RFI CM MAR,A PTR 1 1 Increment
ALU CM PTR 1 F=A minus 1
A
0 Increment
MM CM PC Zero
EIN 1 Zero
DIN 1 Zero
Pre-Skip TEST IND. Conditional
Second
Skip CM A PC Increment
ALU CM PC 1 F=A A
1 Zero
__________________________________________________________________________
Those skilled in the art will recognize that the preferred embodiment described above may be altered and modified without departing from the true spirit and scope of the invention as defined in the accompanying claims.
Claims (8)
1. In a data processor capable of processing data in accordance with a first macroinstruction and a second macroinstruction, improved apparatus for generating at least a first microinstruction and for generating at least a second microinstruction for use in combination with the first microinstruction for executing the second macroinstruction, comprising:
microinstruction memory means for storing the first microinstruction at a first address and the second microinstruction at a second address;
output means for storing the microinstructions read from the microinstruction memory means and for generating a first output signal in response to the absence of the first microinstruction from the output means and for generating a second output signal in response to the presence of the first microinstruction in the output means;
instruction register means for storing during a first time period a first operation code signal corresponding to at least a portion of the first macroinstruction and for storing during a second time period a second operation code signal corresponding to at least a portion of the second macroinstruction; and
addressing means responsive during a first time period to the first operation code signal and the first output signal for reading the first microinstruction from the first address into the output means so that the execution of the first macroinstruction is enabled, and responsive during a second time period later than the first time period to the second operation code signal and the first output signal for reading the first microinstruction from the first address into the output means and responsive to the second operation code signal and the second output signal for reading the second microinstruction from the second address into the output means so that the execution of the second macroinstruction is enabled by the first and second microinstructions, whereby the first microinstruction stored at the first address is used to enable the execution of both the first and second macroinstructions.
2. Apparatus, as claimed in claim 1, wherein the addressing means comprises:
translating means for generating a first address signal corresponding to the first address in response to the combined first operation code signal and first output signal and in response to the combined second operation code signal and first output signal, and for generating a second address signal corresponding to the second address in response to the combined second operation code signal and second output signal;
address register means for storing the first and the second address signals while the first and second microinstructions are being read from the microinstruction memory means; and
means for transmitting the first and second address signals from the translating means to the address register means.
3. Apparatus, as claimed in claim 2, wherein the translating means comprises a diode matrix.
4. Apparatus, as claimed in claim 2, wherein the means for transmitting comprises conductors.
5. Apparatus, as claimed in claim 4, wherein the means for transmitting comprises encoder means for preventing the first and second address signals from being transmitted to the address register means and for transmitting to the address register means a priority address signal corresponding to a data processor instruction having a priority greater than the first and second macroinstructions.
6. In a data processor capable of processing data in accordance with a first macroinstruction and a second macroinstruction, improved apparatus for generating a first microinstruction, a second microinstruction and a third microinstruction used to execute the first macroinstruction and for generating a fourth microinstruction used with the first and second microinstructions to execute the second macroinstruction, comprising:
microinstruction memory means for storing the first microinstruction at a first address, the second microinstruction at a second address, the third microinstruction at a third address and the fourth microinstruction at a fourth address and for reading the microinstructions from the addresses in response to address signals;
output means for storing the microinstructions read from the microinstruction memory means and for sequentially generating a first output signal in response to the absence of the first microinstruction from the output means, for generating a second output signal in response to the presence of the first microinstruction in the output means, and for generating a third output signal in response to the presence of the second microinstruction in the output means;
instruction register means for storing during a first time period a first operation code signal corresponding to at least a portion of the first macroinstruction and for storing during a second time period a second operation code signal corresponding to at least a portion of the second macroinstruction;
translating means for sequentially generating a first address signal corresponding to the first address in response to the combination of the first operation code signal and the first output signal and in response to the combination of the second operation code signal and the first output signal, for generating a third address signal corresponding to the third address in response to the combination of the first operation code signal and the third output signal, and for generating a fourth address signal corresponding to the fourth address in response to the combination of the second operation code signal and the third output signal;
address register means for storing the first address signal and for incrementing the first address signal to form the second address signal in response to the second output signal;
means for conducting said first, third and fourth address signals from the translating means to the address register means; and
means for successively transmitting the first, second and third address signals from the address register means to the microinstruction memory means as the first, second and third output signals are successively produced by the output means during the first time period so that the execution of the first macroinstruction is enabled, and for successively transmitting the first, second and fourth address signals to the microinstruction memory means as the first, second and fourth output signals are successively produced by the output means during the second time period so that the execution of the second macroinstruction is enabled.
7. Apparatus, as claimed in claim 6, wherein the translating means comprises a diode matrix.
8. Apparatus, as claimed in claim 7, wherein the means for conducting comprises means for preventing the first, third and fourth address signals from being transmitted to the address register means and for transmitting to the address register means a priority address signal corresponding to a processor instruction having a priority greater than the first and second macroinstructions.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US05/521,022 US3990054A (en) | 1974-11-05 | 1974-11-05 | Microprogram organization techniques |
| US06/006,969 US4273892A (en) | 1974-11-05 | 1979-01-25 | Preparation of hydrophilic polyolefin fibers for use in papermaking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US05/521,022 US3990054A (en) | 1974-11-05 | 1974-11-05 | Microprogram organization techniques |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| US3990054A true US3990054A (en) | 1976-11-02 |
Family
ID=24075004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US05/521,022 Expired - Lifetime US3990054A (en) | 1974-11-05 | 1974-11-05 | Microprogram organization techniques |
Country Status (1)
| Country | Link |
|---|---|
| US (1) | US3990054A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4118776A (en) * | 1975-07-17 | 1978-10-03 | Nippon Electric Company, Ltd. | Numerically controlled machine comprising a microprogrammable computer operable with microprograms for macroinstructions and for inherent functions of the machine |
| US4131943A (en) * | 1976-06-30 | 1978-12-26 | Tokyo Shibaura Electric Co., Ltd. | Microprogrammed computer employing a decode read only memory (DROM) and a microinstruction read only memory (ROM) |
| US4160289A (en) * | 1976-10-07 | 1979-07-03 | Societa Italiana Telecomunicazioni Siemens S.P.A. | Microprogram-control unit for data processor |
| US4162519A (en) * | 1975-01-20 | 1979-07-24 | Nixdorf Computer Ag | Data processor with address allocation to operations |
| US4173041A (en) * | 1976-05-24 | 1979-10-30 | International Business Machines Corporation | Auxiliary microcontrol mechanism for increasing the number of different control actions in a microprogrammed digital data processor having microwords of fixed length |
| US4197578A (en) * | 1977-01-14 | 1980-04-08 | Hitachi, Ltd. | Microprogram controlled data processing system |
| EP0019392A2 (en) * | 1979-05-21 | 1980-11-26 | Motorola, Inc. | Instruction register sequence decoder for microprogrammed data processor and method |
| US4330823A (en) * | 1978-12-06 | 1982-05-18 | Data General Corporation | High speed compact digital computer system with segmentally stored microinstructions |
| US4342080A (en) * | 1978-11-08 | 1982-07-27 | Data General Corporation | Computer with microcode generator system |
| EP0063458A2 (en) * | 1981-04-13 | 1982-10-27 | Texas Instruments Incorporated | Microcomputer system |
| US4493027A (en) * | 1981-05-22 | 1985-01-08 | Data General Corporation | Method of performing a call operation in a digital data processing system having microcode call and return operations |
| US4528648A (en) * | 1982-07-21 | 1985-07-09 | Sperry Corporation | Memory management system |
| US4633390A (en) * | 1980-04-25 | 1986-12-30 | Tokyo Shibaura Denki Kabushiki Kaisha | Microprogram control system |
| US4750108A (en) * | 1982-12-31 | 1988-06-07 | U.S. Philips Corporation | Data processor unit comprising a control section which comprises an address generator for generating addresses which are composed of characteristic address portions |
| US4763246A (en) * | 1984-08-18 | 1988-08-09 | International Computers Limited | Microprogram control |
| US5046040A (en) * | 1986-01-16 | 1991-09-03 | Kabushiki Kaisha Toshiba | Microprogram control apparatus using don't care bits as part of address bits for common instructions and generating variable control bits |
| US5058007A (en) * | 1987-11-05 | 1991-10-15 | Raytheon Company | Next microinstruction generator in a microprogram control unit |
| US5123096A (en) * | 1987-08-03 | 1992-06-16 | Masahito Matuo | Data processor with addressing mode decoding function |
| US5390307A (en) * | 1990-08-29 | 1995-02-14 | Mitsubishi Denki Kabushiki Kaisha | Apparatus for a multi-data store or load instruction for transferring multiple contiguous storage locations in one transfer operation |
| US5611062A (en) * | 1995-03-31 | 1997-03-11 | International Business Machines Corporation | Specialized millicode instruction for string operations |
| US5621909A (en) * | 1996-03-12 | 1997-04-15 | International Business Machines Corporation | Specialized millicode instruction for range checking |
| US5694587A (en) * | 1995-03-31 | 1997-12-02 | International Business Machines Corporation | Specialized millicode instructions for test PSW validity, load with access test, and character translation assist |
| US5754810A (en) * | 1996-03-12 | 1998-05-19 | International Business Machines Corporation | Specialized millicode instruction for certain decimal operations |
| US20070192568A1 (en) * | 2006-02-03 | 2007-08-16 | Fish Russell H Iii | Thread optimized multiprocessor architecture |
| US20080320277A1 (en) * | 2006-02-03 | 2008-12-25 | Russell H. Fish | Thread Optimized Multiprocessor Architecture |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3325788A (en) * | 1964-12-21 | 1967-06-13 | Ibm | Extrinsically variable microprogram controls |
| US3391394A (en) * | 1965-10-22 | 1968-07-02 | Ibm | Microprogram control for a data processing system |
| US3736563A (en) * | 1970-03-31 | 1973-05-29 | Siemens Ag | Program control unit for a digital data processing installation |
| US3766532A (en) * | 1972-04-28 | 1973-10-16 | Nanodata Corp | Data processing system having two levels of program control |
| US3781823A (en) * | 1972-07-28 | 1973-12-25 | Bell Telephone Labor Inc | Computer control unit capable of dynamically reinterpreting instructions |
| US3800293A (en) * | 1972-12-26 | 1974-03-26 | Ibm | Microprogram control subsystem |
| US3839705A (en) * | 1972-12-14 | 1974-10-01 | Gen Electric | Data processor including microprogram control means |
| US3886523A (en) * | 1973-06-05 | 1975-05-27 | Burroughs Corp | Micro program data processor having parallel instruction flow streams for plural levels of sub instruction sets |
-
1974
- 1974-11-05 US US05/521,022 patent/US3990054A/en not_active Expired - Lifetime
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3325788A (en) * | 1964-12-21 | 1967-06-13 | Ibm | Extrinsically variable microprogram controls |
| US3391394A (en) * | 1965-10-22 | 1968-07-02 | Ibm | Microprogram control for a data processing system |
| US3736563A (en) * | 1970-03-31 | 1973-05-29 | Siemens Ag | Program control unit for a digital data processing installation |
| US3766532A (en) * | 1972-04-28 | 1973-10-16 | Nanodata Corp | Data processing system having two levels of program control |
| US3781823A (en) * | 1972-07-28 | 1973-12-25 | Bell Telephone Labor Inc | Computer control unit capable of dynamically reinterpreting instructions |
| US3839705A (en) * | 1972-12-14 | 1974-10-01 | Gen Electric | Data processor including microprogram control means |
| US3800293A (en) * | 1972-12-26 | 1974-03-26 | Ibm | Microprogram control subsystem |
| US3886523A (en) * | 1973-06-05 | 1975-05-27 | Burroughs Corp | Micro program data processor having parallel instruction flow streams for plural levels of sub instruction sets |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4162519A (en) * | 1975-01-20 | 1979-07-24 | Nixdorf Computer Ag | Data processor with address allocation to operations |
| US4118776A (en) * | 1975-07-17 | 1978-10-03 | Nippon Electric Company, Ltd. | Numerically controlled machine comprising a microprogrammable computer operable with microprograms for macroinstructions and for inherent functions of the machine |
| US4173041A (en) * | 1976-05-24 | 1979-10-30 | International Business Machines Corporation | Auxiliary microcontrol mechanism for increasing the number of different control actions in a microprogrammed digital data processor having microwords of fixed length |
| US4131943A (en) * | 1976-06-30 | 1978-12-26 | Tokyo Shibaura Electric Co., Ltd. | Microprogrammed computer employing a decode read only memory (DROM) and a microinstruction read only memory (ROM) |
| US4160289A (en) * | 1976-10-07 | 1979-07-03 | Societa Italiana Telecomunicazioni Siemens S.P.A. | Microprogram-control unit for data processor |
| US4197578A (en) * | 1977-01-14 | 1980-04-08 | Hitachi, Ltd. | Microprogram controlled data processing system |
| US4342080A (en) * | 1978-11-08 | 1982-07-27 | Data General Corporation | Computer with microcode generator system |
| US4330823A (en) * | 1978-12-06 | 1982-05-18 | Data General Corporation | High speed compact digital computer system with segmentally stored microinstructions |
| EP0019392A3 (en) * | 1979-05-21 | 1981-06-10 | Motorola, Inc. | Instruction register sequence decoder for microprogrammed data processor and method |
| US4342078A (en) * | 1979-05-21 | 1982-07-27 | Motorola, Inc. | Instruction register sequence decoder for microprogrammed data processor and method |
| EP0019392A2 (en) * | 1979-05-21 | 1980-11-26 | Motorola, Inc. | Instruction register sequence decoder for microprogrammed data processor and method |
| US4633390A (en) * | 1980-04-25 | 1986-12-30 | Tokyo Shibaura Denki Kabushiki Kaisha | Microprogram control system |
| EP0063458A2 (en) * | 1981-04-13 | 1982-10-27 | Texas Instruments Incorporated | Microcomputer system |
| EP0063458A3 (en) * | 1981-04-13 | 1985-05-29 | Texas Instruments Incorporated | Microcomputer system |
| US4493027A (en) * | 1981-05-22 | 1985-01-08 | Data General Corporation | Method of performing a call operation in a digital data processing system having microcode call and return operations |
| US4528648A (en) * | 1982-07-21 | 1985-07-09 | Sperry Corporation | Memory management system |
| US4750108A (en) * | 1982-12-31 | 1988-06-07 | U.S. Philips Corporation | Data processor unit comprising a control section which comprises an address generator for generating addresses which are composed of characteristic address portions |
| US4763246A (en) * | 1984-08-18 | 1988-08-09 | International Computers Limited | Microprogram control |
| US5046040A (en) * | 1986-01-16 | 1991-09-03 | Kabushiki Kaisha Toshiba | Microprogram control apparatus using don't care bits as part of address bits for common instructions and generating variable control bits |
| US5123096A (en) * | 1987-08-03 | 1992-06-16 | Masahito Matuo | Data processor with addressing mode decoding function |
| US5058007A (en) * | 1987-11-05 | 1991-10-15 | Raytheon Company | Next microinstruction generator in a microprogram control unit |
| US5390307A (en) * | 1990-08-29 | 1995-02-14 | Mitsubishi Denki Kabushiki Kaisha | Apparatus for a multi-data store or load instruction for transferring multiple contiguous storage locations in one transfer operation |
| US5694587A (en) * | 1995-03-31 | 1997-12-02 | International Business Machines Corporation | Specialized millicode instructions for test PSW validity, load with access test, and character translation assist |
| US5611062A (en) * | 1995-03-31 | 1997-03-11 | International Business Machines Corporation | Specialized millicode instruction for string operations |
| US5621909A (en) * | 1996-03-12 | 1997-04-15 | International Business Machines Corporation | Specialized millicode instruction for range checking |
| US5754810A (en) * | 1996-03-12 | 1998-05-19 | International Business Machines Corporation | Specialized millicode instruction for certain decimal operations |
| US20070192568A1 (en) * | 2006-02-03 | 2007-08-16 | Fish Russell H Iii | Thread optimized multiprocessor architecture |
| US20080320277A1 (en) * | 2006-02-03 | 2008-12-25 | Russell H. Fish | Thread Optimized Multiprocessor Architecture |
| AU2008355072B2 (en) * | 2006-02-03 | 2012-05-31 | Russell H. Fish Iii | Thread optimized multiprocessor architecture |
| AU2008355072C1 (en) * | 2006-02-03 | 2012-11-29 | Russell H. Fish Iii | Thread optimized multiprocessor architecture |
| US8977836B2 (en) * | 2006-02-03 | 2015-03-10 | Russell H. Fish, III | Thread optimized multiprocessor architecture |
| US8984256B2 (en) * | 2006-02-03 | 2015-03-17 | Russell Fish | Thread optimized multiprocessor architecture |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US3990054A (en) | Microprogram organization techniques | |
| US4118773A (en) | Microprogram memory bank addressing system | |
| US3991404A (en) | Apparatus for carrying out macroinstructions in a microprogrammed computer | |
| US4042972A (en) | Microprogram data processing technique and apparatus | |
| US4631663A (en) | Macroinstruction execution in a microprogram-controlled processor | |
| US4168523A (en) | Data processor utilizing a two level microaddressing controller | |
| US3949370A (en) | Programmable logic array control section for data processing system | |
| US3990052A (en) | Central processing unit employing microprogrammable control for use in a data processing system | |
| US3739352A (en) | Variable word width processor control | |
| EP0114304B1 (en) | Vector processing hardware assist and method | |
| US3983539A (en) | Polymorphic programmable units employing plural levels of sub-instruction sets | |
| EP0011442B1 (en) | Data processing system having an integrated stack and register machine architecture | |
| US4001788A (en) | Pathfinder microprogram control system | |
| US4179731A (en) | Microprogrammed control system | |
| US4075687A (en) | Microprogram controlled digital computer | |
| US3585605A (en) | Associative memory data processor | |
| US4156925A (en) | Overlapped and interleaved control store with address modifiers | |
| US3983541A (en) | Polymorphic programmable units employing plural levels of phased sub-instruction sets | |
| US4124893A (en) | Microword address branching bit arrangement | |
| US4323964A (en) | CPU Employing micro programmable control for use in a data processing system | |
| US4388682A (en) | Microprogrammable instruction translator | |
| US3958221A (en) | Method and apparatus for locating effective operand of an instruction | |
| US4370729A (en) | Microprogram sequencer | |
| EP0126125B1 (en) | Multiple control stores for a pipelined microcontroller | |
| CA1099415A (en) | Rom initializing apparatus |